Regex to Remove HTML Tags — Strip Markup, Keep Text

Stripping HTML Tags with Regex

Removing HTML tags from a string is a common need for plain-text previews, search index input, and email subject lines. For untrusted input always use DOMPurify plus a whitelist; for trusted input regex is fast and dependency-free.

Basic Strip

str.replace(/<[^>]+>/g, "")

Removes any tag-shaped substring.

Strip Tags AND Comments

str.replace(/<!--[\s\S]*?-->/g, "").replace(/<[^>]+>/g, "")

The first pass removes comments (which can span multiple lines), the second removes tags.

Strip Tags but Keep Whitespace Where They Were

To turn <p>hello</p><p>world</p> into hello world (with a space) instead of helloworld:

str.replace(/<\/?(?:p|div|br|li|h[1-6])[^>]*>/gi, " ")
   .replace(/<[^>]+>/g, "")
   .replace(/\s+/g, " ")
   .trim()

Tested Examples

Input	Output
<p>hello</p>	hello
<a href="x">link</a>	link
<!-- comment -->visible	visible (with comment-strip)
<p>a</p><p>b</p>	ab (basic) / a b (block-aware)
<script>alert(1)</script>	`` (basic strips tag, leaves alert(1))

Why You Still Want a Real Parser

Regex strip leaves behind:

• Script and style content (alert(1) survives the strip)
• HTML entities (& is unchanged)
• Malformed input that can be exploited

To handle script content, strip those blocks first:

str.replace(/<(script|style)[^>]*>[\s\S]*?<\/\1>/gi, "")

Decode Entities

After stripping tags, decode entities to get readable text:

const txt = document.createElement("textarea");
txt.innerHTML = stripped;
const decoded = txt.value;

Recommendation

For server-rendered preview text, the script/style strip plus tag strip plus entity decode is sufficient. For untrusted user input that will be re-rendered, never trust regex; sanitize with a battle-tested library.